- Email: [email protected]
Virulence of Mycobacterium avium complex strains isolated from immunocompetent patients
Microbial Pathogenesis 46 (2009) 6–12
Contents lists available at ScienceDirect
Microbial Pathogenesis journal homepage: www.elsevier.com/locate/micpath
Virulence of Mycobacterium avium complex strains isolated from immunocompetent patients Yoshitaka Tateishi a, b, *, Yukio Hirayama a, Yuriko Ozeki a, c, Yukiko Nishiuchi a, d, Mamiko Yoshimura a, Jing Kang a, Atsushi Shibata a, Kazuto Hirata e, Seigo Kitada b, Ryoji Maekura b, Hisashi Ogura f, Kazuo Kobayashi g, Sohkichi Matsumoto a, * a
Department of Bacteriology, Osaka City University Graduate School of Medicine, 1-4-3 Asahi-machi, Abeno-ku, Osaka 545-8585, Japan Department of Internal Medicine, National Hospital Organization Toneyama National Hospital, 5-1-1 Toneyama, Toyonaka, Osaka 560-8552, Japan Sonoda Women’s University, 7-29-1 Minamitsukaguchi-cho, Amagasaki, Hyogo 661-8520, Japan d Toneyama Institute for Tuberculosis Research, Osaka City University Medical School, Toyonaka 5-1-1, Toneyama, Toyonaka, Osaka 560-8552, Japan e Department of Respiratory Medicine, Osaka City University Graduate School of Medicine, 1-4-3 Asahi-machi, Abeno-ku, Osaka 545-8585, Japan f Department of Virology, Osaka City University Graduate School of Medicine, 1-4-3 Asahi-machi, Abeno-ku, Osaka 545-8585, Japan g Department of Immunology, National Institute of Infectious Diseases, 1-23-1, Toyama, Shinjuku-ku, Tokyo 162-8640, Japan b c
a r t i c l e i n f o
a b s t r a c t
Article history: Received 6 August 2008 Received in revised form 29 September 2008 Accepted 2 October 2008 Available online 1 November 2008
Mycobacterium avium complex (MAC) disease has been increasing worldwide not only in immunocompromised but also in immunocompetent humans. However, the relationship between mycobacterial strain virulence and disease progression in immunocompetent humans is unclear. In this study, we isolated 6 strains from patients with pulmonary MAC disease. To explore the virulence, we examined the growth in human THP-1 macrophages and pathogenicity in C57BL/6 mice. We found that one strain, designated 198, which was isolated from a patient showing the most progressive disease, persisted in THP-1 cells. In addition, strain 198 grew to a high bacterial load with strong inflammation in mouse lungs and spleens 16 weeks after infection. To our knowledge, strain 198 is the first isolated MAC strain that exhibits hypervirulence consistently for the human patient, human macrophages in vitro, and even for immunocompetent mice. Other strains showed limited survival and weak virulence both in macrophages and in mice, uncorrelated to disease progression in human patients. We demonstrated that there is a hypervirulent clinical MAC strain whose experimental virulence corresponds to the serious disease progression in the patients. The existence of such strain suggests the involvement of bacterial virulence in the pathogenesis of pulmonary MAC disease in immunocompetent status. Ó 2008 Elsevier Ltd. All rights reserved.
Keywords: Mycobacterium avium complex Virulence Clinical isolates Immunocompetent humans Pulmonary disease
1. Introduction Mycobacterium avium complex (MAC) is the most common cause of human infection due to nontuberculous mycobacteria. Initially MAC was regarded as only an opportunistic pathogen, primarily in acquired immunodeficiency syndrome (AIDS) patients [1]; however, it has now been shown to cause progressive pulmonary disease even in immunocompetent humans [2]. The American Thoracic Society indicates a wide range of clinical manifestation in patients with non-AIDS MAC disease; some patients keep a stable condition for years, whereas others progress their illness rapidly [3]. Furthermore, MAC infection can be more difficult to treat
* Corresponding authors. Department of Bacteriology, Osaka City University, Graduate School of Medicine, 1-4-3 Asahi-machi, Abeno-ku, Osaka 545-8585, Japan. Tel.: þ81 6 6645 3746; fax: þ81 6 6645 3747. E-mail address: [email protected] (Y. Tateishi). 0882-4010/$ – see front matter Ó 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.micpath.2008.10.007
than M. tuberculosis due to even fewer available anti-microbial agents [3]. The pathogenesis of MAC infection has been recently investigated with respect to the host immune response. Interferongamma (IFN-g) activates macrophages to produce proteolytic enzymes and other metabolites, which exhibit mycobactericidal effects. Tumor necrosis factor-alpha (TNF-a), of which production is also stimulated by IFN-g, augments the bactericidal capacity of macrophages and plays a key role in the induction of the acquired immune response against mycobacteria [4]. A defective IFN-g response has been shown recently to cause disseminated MAC disease in IFN-g knock out mice and in humans with genetic mutations of IFN-g receptor [5,6] or autoantibodies to IFN-g in some young non-AIDS patients [7,8]. In addition to that, the activity of interleukin-10 (IL-10), which is known to inhibit cytokine synthesis by IFN-g-producing type1 helper T cells (Th1 cells), has been shown to increase susceptibility to MAC infection in immunocompetent mice [9].
Y. Tateishi et al. / Microbial Pathogenesis 46 (2009) 6–12
Besides genetic factors of the host, bacterial virulence should play an important role for the development of MAC disease. While isolates of M. tuberculosis are genetically homogeneous at the nucleotide level [10], MAC has high genetic diversity, including the presence of multiple plasmids [11], and thus likely to have a large corresponding diversity in virulence. In the most complete study examining virulence, forty-one MAC isolates from the environment as well as infected humans and animals were compared for virulence in C57BL/6 mice by intravenous injection [12]. Monitoring of the virulence by CFU counts in lungs, livers, and spleens over 4 months revealed three virulence phenotypes; high (logarithmically increasing load), intermediate (chronic infection at a constant load), and low (initial load increase followed by a decrease until clearance). In addition, clinical studies have suggested severe disease outcome in patients infected with some specific strain type of MAC. For example, MAC serovars 1, 4, and 8 Mycobacterium avium are associated with disease severity in AIDS patients [13], and a serovar 4 M. avium isolate from an AIDS patient was more invasive and proliferative in blood mononuclear cell-derived human macrophages than a serovar 2 strain from chickens [14]. In non-AIDS MAC disease, Mycobacterium intracellulare is associated with greater disease progression [15], and moreover, our previous prospective study on 68 non-AIDS patients suggests that serovar 4 M. avium is linked to greater disease progression with a pulmonary MAC infection [16]. Taking these previous data into consideration, we hypothesize that relatively hypervirulent MAC strains exist and may be associated with serious disease progression in immunocompetent patients. In order to elucidate the involvement of mycobacterial virulence in the pathogenesis of human pulmonary MAC disease, in this study we examined the difference of mycobacterial virulence of clinical isolates from patients with different disease types using human macrophages and immunocompetent mice.
7
and numerous bacteria in the sputum. The patient infected with strain 198 exhibited the most serious disease outcome among study patients in that a right pneumonectomy was needed to prevent disease progression. Strains 27, 36, and 347 were derived from patients with little progression of disease without chemotherapy (silent type). They exhibited lower levels of ESR, segmental pulmonary lesions in chest X-ray findings, and fewer bacteria in the sputum. The isolates belonging to the progressive type consisted of M. intracellulare unclassified serovar similar to serovar 12 (strain 198) and M. avium apolar type (strains 33 and 288). The isolates belonging to the silent type consisted of M. intracellulare serovar 1 (strain 27) and M. avium apolar type (strains 36 and 347). For comparison, we employed 2 veterinary strains of M. avium ATCC 25291 (serovar 2) as a highly virulent strain in mice [12] and ATCC 35767 (serovar 4) as a low virulent strain. Four clinical isolates other than strains 33 and 347, and ATCC 25291 formed the transparent colony morphology. Strain 33 produced both transparent and rough colony morphologies. Strain 347 and ATCC 35767 displayed smooth opaque colony morphology.
2.2. Growth of clinical isolates in 7H9 broth All strains showed logarithmic growth from 3 days after culture in 7H9 broth (Table 2). At day 5, two isolates from progressive type (strains 198 and 288) and one isolate from silent type (strain 36) grew significantly slower than ATCC 25291 (P < 0.005), and all clinical strains grew significantly slower than ATCC 35767 (P < 0.0001). The growth of strain 198 at day 5 was significantly slower than that of strain 27 (P ¼ 0.001), and was not significantly different from that of other clinical isolates.
2.3. Virulence of clinical isolates in THP-1 monocyte-derived macrophages
2. Results 2.1. Characteristics of mycobacterial strains Six clinical isolates of MAC were isolated from sputum of nonAIDS patients with pulmonary MAC disease, and designated 27, 33, 36, 198, 288, and 347 (Table 1). Strains 33, 198 and 288 were derived from patients with progressive disease against combination chemotherapy recommended by the American Thoracic Society guideline (progressive type) [3]. The patients with progressive disease exhibited higher levels of erythrocyte sedimentation rate (ESR), diffuse and severe pulmonary lesions in chest X-ray findings,
We next studied intracellular survival of the isolates. THP-1 cells, a human monocytic cell line, were differentiated into macrophages by treatment with phorbol 12-myristate 13-acetate (PMA) and infected with MAC strains. Strain 198 grew in THP-1 cells significantly higher than any other strains during 7 days of infection (P < 0.0001) (Table 3). Strain 198 grew to approximately 20-fold during 2 days of infection (P ¼ 0.005), and even at day 7, it kept the same level of bacterial load as day 0. Strain 36 also grew to approximately 2-fold during 2 days of infection (P ¼ 0.008); however, it was rapidly eliminated at day 7, similar to the other strains except for strain 198. There was no significant difference in
Table 1 Characteristics of isolated strains and clinical findings. Isolates
Species and serovar
Age
Sex
Duration of illness (years)
Erythrocyte sedimentation rate (mm/h)
Chest X-ray findingsa
Sputumb Smear
Culture
Progressive type 33 M. avium apolar type 198 M. intracellulare unclassified serovarc 288 M. avium apolar type
58 62 56
M F F
17 3 12
62 108 78
Advanced Advanced Advanced
2þ 2þ 2þ
3þ 2þ 2þ
Silent type 27 36 347
67 54 79
F F F
17 9 14
50 29 50
Moderate Moderate Moderate
– – 1þ
1þ 1þ 1þ
M. intracellulare serovar 1 M. avium apolar type M. avium apolar type
Data and sputum samples were collected at the enrollment of the study in 2003. a Advanced chest X-ray findings were defined as bilateral cavities, giant cavities, or bilateral bronchiectasis, and moderate findings were defined as focal inflammation, small or fewer cavities, or mild bronchiectasis. b Smear findings of sputum were defined as follows in high performance fields of microscopy; -: no bacteria in all fields, 1þ: less than one bacteria in several fields, 2þ: approximately 1–12 bacteria in one field. Culture findings were defined as follows using Ogawa egg agar; 1þ: colonies less than 200, 2þ: colonies more than 200 and less than 500, 3þ: colonies more than 500 and less than 2000. c The serovar of strain 198 was identified as a new type similar to serovar 12 determined by the liquid chromatography/mass spectrometry.
8
Y. Tateishi et al. / Microbial Pathogenesis 46 (2009) 6–12
Table 2 Growth rate of MAC in 7H9 broth. Strain
33 198 288 27 36 347 25291 35767
Ratio of CFUs ata Day 1
Day 3
Day 5
0.91 0.28 0.93 0.17 0.85 0.20 1.1 0.25 0.83 0.093 0.96 0.16 1.6 0.25 0.97 0.12
38 0.86 12 1.9 4.3 1.9 7.3 2.2 5.4 0.21 2.7 1.1 4.8 0.24 11 3.0
63 10 18 4.0 16 0.95 120 21 11.0 1.7 44 19 110 16* 370 43**
and spleens, and severe pulmonary inflammation at 16 weeks of infection. By contrast, other clinical isolates did not increase profoundly in lung CFUs; however, these strains were never eliminated from lungs. ATCC 35767 was rapidly decreased and undetectable in lungs, spleens and livers within 16 weeks of infection. Overall, the clinical isolates other than strain 198 exhibited limited histological lesions with transient inflammatory changes in lungs 4 weeks after the inoculation, and thereafter the inflammation subsided at 16 weeks. 3. Discussion
*
Significantly different (P < 0.005) from values for strain 198, 288, and 36 as calculated by Scheffe´’s test. ** Significantly different (P < 0.0001) from values for all clinical strains as calculated by Scheffe´’s test. a Means standard deviations of the ratio of CFUs to those at day 0.
the growth rate among these strains except strain 198 during infection. On light microscopic observation, THP-1 cell morphologies were not different between infected and uninfected cells (data not shown). We then assessed cytotoxicity by the levels of lactate dehydrogenase (LDH) released into the culture supernatants at day 7. The LDH release was detectable in strain 33 and the laboratory strains (strain 33; 5.8 1.5%, ATCC 25291; 11 1.0%, ATCC 35767; 12 1.9%, without significant difference among these strains); however, it was not detectable in other clinical isolates.
2.4. Pathogenesis of clinical isolates in mice Female C57BL/6 mice were infected by intratracheal instillation with each strain. Bacterial load in lungs, livers, and spleens were evaluated, and histological inflammation was visually analyzed in 5-mice per strain at defined time points during 16 weeks of infection. There was no significant difference in lung CFUs among strains tested 1 day after the inoculation. Strain 198 showed high bacterial load, and tended to increase gradually both in lungs and spleens during 16 weeks of infection (P ¼ 0.08 between day 1 and 16 weeks) (Fig. 1). Strain 198 was loaded in lungs significantly higher than strain 27 (P ¼ 0.04) and 33 (P ¼ 0.0006) at 8 weeks of infection, and than strain 33 (P ¼ 0.0009), 288 (P ¼ 0.001), 36 (P ¼ 0.0003), and 347 (P ¼ 0.004) at 16 weeks. Histologically, strain 198 induced strong inflammation in lungs, which was paralleled with bacterial loads (Fig. 2). ATCC 25291, known as highly a virulent strain in mice [12], showed initial reduction of bacterial load in lungs at 4 weeks of infection (P ¼ 0.01 between day 1 and 4 weeks) and rapid increase in bacterial load in lungs after 4 weeks of infection. ATCC 25291 was comparatively virulent to strain 198 with respect to the high bacterial load in lungs
Table 3 Growth rate of MAC in THP-1 cells. strain
33 198 288 27 36 347 25291 35767 *
Ratio of CFUs ata Day 2
Day 7
0.29 0.12 18 12* 0.47 0.26 0.54 0.37 2.4 1.2 1.1 0.49 0.16 0.048 0.059 0.029
0.25 0.16 1.30 0.68* 0.097 0.055 0.28 0.11 0.31 0.11 0.36 0.23 0.11 0.038 0.0070 0.0048
Significantly different (P < 0.0001) from values for any other strains studied as calculated by Scheffe´’s test. a Means standard deviations of the ratio of CFUs to those at day 0.
Virulence is defined as the quantitative ability of an agent to cause disease. The virulence of mycobacteria can be evaluated by the infection to macrophages and animals [17]. This is the first study that examined the virulence of MAC isolates from immunocompetent patients with different types of disease outcome. We found that strain 198, which derived from a patient with most serious disease, revealed high bacterial load both in THP-1 cells and in C57BL/6 mice among isolates studied. Strain-specific virulence of MAC has been implicated by some previous studies of the serovar 4 M. avium isolated from patients. In AIDS-related MAC disease, a serovar 4 M. avium isolate has shown to be one of the frequently isolated type [13], and a previous analysis of a serovar 4 isolate and ATCC strain has shown the superior virulence of serovar 4 M. avium in human macrophages [14]. In non-AIDS pulmonary MAC disease, our recent prospective study indicates that patients infected with serovar 4 M. avium has poorer prognosis than those infected with MAC of other serovars [16]; however, to our best knowledge, no study has shown the direct data of mycobacterial virulence of clinical isolates and clinical disease outcome. Strain 198 is the first MAC isolate whose experimental virulence corresponds to the serious disease outcome in humans. Thus, strain 198 has strainspecific strong virulence for immunocompetent humans and mice. We consider that strain 198 is worth further genetic investigation of virulence factors. MAC strains hypervirulent for mice has been isolated previously by Pedrosa et al. including ATCC 25291 and MAC 101, which proliferate profoundly in mouse macrophages and in mice in vivo [12]. In this study, strain 198 proliferated in human macrophages, in correspondence with rapid clinical disease progression and additionally in mouse lungs. The consistency between experimental virulence in human cells and clinical disease outcome suggests that the capability of inducing such strong pathogenesis may be attributed mostly to the characteristics of the pathogen, i.e. virulence factor(s) for mammalian cells unique to strain 198. Previously Birkness et al. has shown the strong cytotoxic effect and growth of serovar 4 M. avium isolated from an AIDS patient in blood mononuclear cell-derived human macrophages compared with a serovar 2 strain from chickens (ATCC 35713). Therefore, we evaluated cytotoxicity of MAC strains by the microscopic morphology and by the LDH release from infected THP-1 cells; however, contrary to the expectation, strain 198 was not cytotoxic to THP-1. In addition, the release of LDH was lower in strain 33, ATCC 25291, and ATCC 35767 than in the previous experiment of M. tuberculosis infection to THP1 cells (cytotoxicity in cases of M. tuberculosis H37Rv and H37Ra; approximately 30%) [18]. We assume that cytotoxic effect may not play a major role in displaying the virulence of MAC during infection, suggested by the similar result by Huttunen et al. showing the lack of cytotoxic effect of MAC in human 28SC macrophage and A549 lung epithelial cell lines evaluated by 3-(4,5-dimethyl-2thiazolyl)-2,5-diphenyl-2H tetrazolium bromide (MTT) assay [19]. We speculate the virulence of strain 198 depends on the ability to survive or proliferate in macrophages rather than cytotoxic effect, which, may be causative for severe pulmonary MAC disease with rapid disease progression within a few years.
Y. Tateishi et al. / Microbial Pathogenesis 46 (2009) 6–12
CFUs/organ
109
33
109 108
108
107
107
107
106
106
106
105
105
105
104
104
104
103
103
103
102
102
102
101
101
101
109
CFUs/organ
109
108
1day 4wks 8wks 16wks
1day 4wks 8wks 16wks
27
109
36
109
108
108
107
107
107
106
106
106
105
105
105
104
104
104
103
103
103
102
102
102
101
101
101
109
25291
1day 4wks 8wks 16wks 109
108
108
107
107
106
106
105
105
104
104
103
103
102
102
101
101 1day 4wks 8wks 16wks
288
1day 4wks 8wks 16wks
108
1day 4wks 8wks 16wks
CFUs/organ
198
9
347
1day 4wks 8wks 16wks
35767 Lung Liver Spleen
1day 4wks 8wks 16wks
Fig. 1. Time course of mycobacterial growth in lungs, spleens and livers of C57BL/6 mice. Bacterial suspensions containing 1 105 CFUs were inoculated intratracheally to female C57BL/6 mice at the age of 7 weeks (n ¼ 20 per strain). The lungs, livers and spleens of 5 mice per strain were sectioned at day 1 (only lungs), 4, 8, and 16 weeks later from challenge Data were presented as means standard deviations of CFUs/organ.
In this study, strain 198 showed strong virulence for mice at 16 weeks of infection similar to ATCC 25291; however, virulence for THP-1 cells was quite different, and pathogenic effects for mice within 4 weeks of infection was dissimilar between these two strains. These differences can be explained by the difference of immune response between host species and by the difference of immune phase. First, strain 198 could proliferate, but ATCC 25291 was rapidly eliminated in THP-1 cells (Table 3). In mouse macrophages mycobactericidal activity is attributed to nitric oxide produced by inducible nitric oxide synthase [20], whereas in human macrophages, it is attributed to Toll-like receptor signalingdependent production of anti-microbial peptides [21,22]. Strain 198 is capable of proliferating under these two patterns of mycobactericidal activities, which suggests that strain 198 may have some virulence factors advantageous to survive both in human and
mouse macrophages against mycobactericidal activity of the hosts, in contrast to ATCC 25291 which may lack virulence factors to survive in human macrophages. Second, strain 198 showed high bacterial load in lungs continuingly during 16 weeks of infection in mice, while ATCC 25291 proliferated after initial reduction in lungs at 4 weeks of infection (Fig. 1). The in vitro infection model using cell lines and the in vivo infection model using mice within 4 weeks reflects early stages of infection; on the other hand, the in vivo model after 8 weeks reflects chronic phase of infection [17,23]. The difference of pathogenic effects for mice within 4 weeks of infection suggests that strain 198 may resist both innate and acquired immunity, while ATCC 25291 may resist acquired immunity only. We assume that strain 198 and ATCC 25291 may possess different virulence mechanisms to persist in in vivo after development of acquired immunity.
10
Y. Tateishi et al. / Microbial Pathogenesis 46 (2009) 6–12
Fig. 2. Histological pictures of the lungs during 4 weeks or 16 weeks of infection in C57BL/6 mice by hematoxylin–eosin staining. Magnification, 40.
In this study, clinical strains except for strain 198 did not show consistent virulence-associated phenotype among THP-1 cells, C57BL/6 mice, and clinical disease outcome. Similarly, Pedrosa et al. has also revealed that the growth of MAC in bone-marrow derived macrophages does not necessarily predict the virulence in mice by comparing the growth of 41 MAC isolates from various derives including humans, animals, and environment [12]. Although some clinical cases of pulmonary MAC disease may be caused by hypervirulent strains such as strain 198, these findings of clinical and
natural isolates suggest that virulence may not the only determinant of the pathogenesis of pulmonary MAC disease in the majority of clinical cases. The development of pulmonary MAC disease depends on the balance between bacterial virulence and host defense. It is widely accepted that patients with pulmonary MAC disease have some characteristics of clinical background, such as males in their 40s and early 50s who have a history of cigarette smoking and excessive alcohol use, and such as postmenopausal, nonsmoking females [3], and these patient characteristics might
Y. Tateishi et al. / Microbial Pathogenesis 46 (2009) 6–12
possibly indicate unknown predisposing conditions which enhance susceptibility for pulmonary MAC infection. The diverse phenotype of clinical MAC strains may be attributed to the disease susceptibility of the hosts. We propose that pathogenic mechanism of human pulmonary MAC disease include two patterns; one is that the strong virulence of MAC strains such as strain 198 induces rapid mycobacterial growth and serious disease outcome, and the other is that relatively weak to moderate virulence interacts with predisposing conditions of the host, leading the wide range of clinical outcome. This study was preliminary in that we did not identify the mechanism of hypervirulence of strain 198. We observed the consistency between hypervirulence in human macrophages bedsides in immunocompetent mice and severe clinical outcome only in strain 198, not in any other isolates studied. From this finding, we speculate the existence of strain-specific virulence factors of strain 198. Recent exponential advances have enabled whole genome sequence of two M. avium strains, M. avium 104 and M. avium subsp. paratuberculosis K-10. Based on these exhaustive information, comparative genomics of MAC organisms has revealed the different genomic components regarding virulence factors, such as ser2 encoding glycosylation enzyme of the lipopeptide core to generate the glycopeptidolipids, mammalian cell entry (mce) gene homologs, and PE/PPE genes (i.e., with Pro Glu and Pro Pro Glu motifs) [24]. In addition, there are large sequence polymorphisms among MAC organisms, suggesting a large corresponding diversity in virulence [11,24]. We speculate that the virulence of MAC strains including strain 198 may be determined by insertion or deletion of virulence genes encoding known [24] or unknown virulence factors. In summary, we demonstrated that certain clinical strain derived from patients of the progressive pulmonary MAC disease exhibits strong virulence in human macrophages and in immunocompetent mice. Among clinical isolates, strain 198 is the first isolate hypervirulent to both human macrophages and mice. Our data suggest that strain-to-strain differences in virulence may play a significant role in disease progression in humans. Although Sarmento et al. showed that capability of TNF-a production from macrophages inversely correlates with the virulence of MAC strains [25], we could not find such relationship among the isolates (data not shown). In future studies, we will identify the virulence/pathogenicity-associated factor(s) of strain 198 and survey the frequency of strain variation in immunocompetent patients with pulmonary MAC disease. 4. Materials and methods 4.1. Bacterial strains We used six clinical isolates from non-AIDS patients with pulmonary MAC disease and two laboratory strains, M. avium ATCC 25291 (serovar 2) and M. avium ATCC 35767 (serovar 4), in this study. Clinical isolates were obtained between September and November in 2003 at Toneyama National Hospital. Informed consent was obtained from all patients according to the guideline of Institutional Review Board of Toneyama National Hospital. Diagnosis of pulmonary MAC disease was made according to the American Thoracic Society guideline [3]. The samples were derived from two groups of patients; one group exhibited progressive disease in spite of the combination chemotherapy including clarithromycin, ethambutol and rifampin recommended by the American Thoracic Society guideline (progressive type) [3], the other displayed no exacerbation without anti-microbial chemotherapy for approximately ten years or more (silent type). These types were determined by the laboratory findings at the period of sputum sampling (including sputum smear and culture,
11
chest X-ray findings, and erythrocyte sedimentation rate) and the rapidness of disease progression (Table 1). Sputum specimens were mixed with 2% sodium hydroxide, and N-acetyl-L-cysteine and then centrifuged for 15 min at 3000 g. The supernatants were discarded, and the sediment was mixed at 1:10 (vol/vol) with sterile water. The bacteria were cultivated in Middlebrook 7H9 broth supplemented with albumin–dextrose–catalase, 0.02% glycerin and 0.05% Tween 80, and then kept at 80 C until following experiments. Identification of MAC was made by polymerase chain reaction using a commercially available kit (AMPLICOR Mycobacterium Tuberculosis Test, Roche, Basel, Switzerland). The serovars of clinical isolates were identified by the liquid chromatography/mass spectrometry as described previously [26]. Strains not containing serovar-specific oligosaccharides were defined as apolar type.
4.2. Growth in 7H9 broth Bacterial suspension was adjusted to be 0.2 by optical density (OD) at 630 nm. The samples were cultured in 5 ml of 7H9 media in plastic tubes without agitation After vortexing to dissolve aggregates, cultivated bacterial suspensions were inoculated at days 1, 3, and 5 by serial 10-fold dilutions on Middlebrook 7H11 agar plates supplemented with oleic acid–albumin–dextrose–catalase, and 0.05% glycerol (7H11-OADC) agar plates in triplicate. The number of CFUs was counted after cultivating at 37 C for 3 weeks. 4.3. Infection of THP-1 cells with MAC in vitro THP-1 cells were purchased from Health Science Research Resources Bank (Tokyo, Japan). The cells were cultured in RPMI1640 containing 10% heat-inactivated fetal bovine serum (FBS; Equitech-bio, TX), and subcultured every 3–4 days. THP-1 cells were differentiated by 100 nM PMA (Sigma–Aldrich, St Louis, MO) for 48 h before infection. Before 48 h of infection, 1 ml of 2 105/ml cells was cultured in RPMI1640 containing 5% human serum (ABblood group) in 24-well plates. Then, 1 ml of 2 104 CFUs/ml bacteria was exposed to the cultured cells for 24 h without opsonization (multiplicity of infection; 0.1 bacteria/cell). After that, the cells were treated with 20 mg/ml of gentamicin for 3 h to kill extracellular bacteria, followed by washing 4 times by RPMI1640. The infected cells were cultured in 2 ml of RPMI1640 containing 5% human serum. At days 0 and 7, uninfected bacteria were removed by washing with RPMI1640 4 times, and 500 ml of filter-sterilized phosphate buffered saline containing 0.5% Triton X-100 (Wako, Osaka, Japan) was treated per well to lyse cell membrane. The intracellular survival of bacteria was determined by counting CFUs by inoculating the cell lysate on 7H11-OADC agar plates. The experiment was performed in triplicate.
4.4. Assays for cytotoxicity Cytotoxic effects were evaluated by the release of LDH from the cells. LDH activity of culture supernatants was determined by a commercially available kit (Roche, Basel, Switzerland). Supernatants were diluted to be 101 by distilled water for optimal reaction. The diluents were reacted with reaction mixture for 30 min, and then the OD was measured at 492 nm. Supernatants of completely lysed uninfected cells with filter-sterilized phosphate buffered saline containing 20% Triton X-100 and those of uninfected cells untreated with Triton X-100 were served as high and low controls, respectively. Cytotoxicity (%) was calculated as follows; ðODsample ODlow control Þ 100=ðODhigh control ODlow control Þ. The measurement was performed in triplicate.
12
Y. Tateishi et al. / Microbial Pathogenesis 46 (2009) 6–12
4.5. Animal studies Female C57BL/6 mice aged at 6 weeks were purchased from CLEA Japan (Tokyo, Japan). All mice were kept under specific pathogen free conditions in animal facility of Osaka City University Graduate School of Medicine according to the institutional guidelines for the animal experiments. Twenty mice were used per group for infecting with each strain. One hundred microlitre of bacterial suspension containing 1 105 CFUs of MAC was inoculated into the trachea of the 7 weeks-aged mice anesthetized with pentobarbital sodium. Lungs, spleens and livers were removed on day 1 (only lungs) and 4, 8, 16 weeks after inoculation from 5-mice per strain. The organs were homogenized in 1 ml saline, and 0.1 ml of 10-fold dilutions of the homogenates was plated on 7H11-OADC agar followed by cultivating for 3 weeks. Bacterial burden was evaluated by CFUs per organ. Histological sections were made by standard methods including formalin fixation, dehydration, embedding in paraffin, and staining with hematoxylin and eosin. 4.6. Statistical analysis Data were analyzed using the statistical analysis software package StatView 5.0 (SAS Institute, Cary, NC). The difference of mycobacterial growth in 7H9 broth, THP-1 cells, and mice was compared by a post hoc test of Scheffe´ among the strains tested. The difference of mycobacterial growth at defined time points during infection in THP-1 cells as well as in mice was compared by repeated measurement ANOVA with a post hoc test of Scheffe´ in the individual strains. Difference was considered statistically significant at P < 0.05. Acknowledgement We thank Todd P. Primm for the critical comments on the manuscript. We also thank Chihiro Inoue and Sara Matsumoto for the assistance of the experiments and for the heartfelt encouragement. This work was supported by grants from the Ministry of Education, Culture, Sports, Science and Technology, the Ministry of Health, Labour and Welfare (Research on Emerging and Reemerging Infectious Diseases, Health Sciences Research Grants), The Japan Health Sciences Foundation, and The United States-Japan Cooperative Medical Science Program against Tuberculosis and Leprosy. References [1] Horsburgh Jr CR, Gettings J, Alexander LN, Lennox JL. Disseminated Mycobacterium avium complex disease among patients infected with human immunodeficiency virus, 1985–2000. Clin Infect Dis 2001;33:1938–43. [2] Field SK, Fisher D, Cowie RL. Mycobacterium avium complex pulmonary disease in patients without HIV infection. Chest 2004;126:566–81. [3] Griffith DE, Aksamit T, Brown-Elliott BA, Catanzaro A, Daley C, Gordin F, et al. An official ATS/IDSA statement: diagnosis, treatment, and prevention of nontuberculous mycobacterial diseases. Am J Respir Crit Care Med 2007;175: 367–416.
[4] Flynn JL, Goldstein MM, Chan J, Triebold KJ, Pfeffer K, Lowenstein CJ, et al. Tumor necrosis factor-a is required in the protective immune response against Mycobacterium tuberculosis in mice. Immunity 1995;2:561–72. [5] Dorman SE, Picard C, Lammas D, Heyne K, van Dissel JT, Baretto R, et al. Clinical features of dominant and recessive interferon g receptor 1 deficiencies. Lancet 2004;364:2113–21. [6] Newport MJ, Huxley CM, Huston S, Hawrylowicz CM, Oostra BA, Williamson R, et al. A mutation in the interferon-g-receptor gene and susceptibility to mycobacterial infection. N Engl J Med 1996;335:1941–9. [7] Kampmann B, Hemingway C, Stephens A, Davidson R, Goodsall A, Anderson S, et al. Acquired predisposition to mycobacterial disease due to autoantibodies to IFN-g. J Clin Invest 2005;115:2480–8. [8] Patel SY, Ding L, Brown MR, Lantz L, Gay T, Cohen S, et al. Anti-IFN-g autoantibodies in disseminated nontuberculous mycobacterial infections. J Immunol 2005;175:4769–76. [9] Roque S, Nobrega C, Appelberg R, Correia-Neves M. IL-10 underlies distinct susceptibility of BALB/c and C57BL/6 mice to Mycobacterium avium infection and influences efficacy of antibiotic therapy. J Immunol 2007;178:8028–35. [10] Ernst JD, Trevejo-Nunez G, Banaiee N. Genomics and the evolution, pathogenesis, and diagnosis of tuberculosis. J Clin Invest 2007;117:1738–45. [11] Primm TP, Lucero CA, Falkinham 3rd JO. Health impacts of environmental mycobacteria. Clin Microbiol Rev 2004;17:98–106. [12] Pedrosa J, Florido M, Kunze ZM, Castro AG, Portaels F, McFadden J, et al. Characterization of the virulence of Mycobacterium avium complex (MAC) isolates in mice. Clin Exp Immunol 1994;98:210–6. [13] Hoffner SE, Kallenius G, Petrini B, Brennan PJ, Tsang AY. Serovars of Mycobacterium avium complex isolated from patients in Sweden. J Clin Microbiol 1990;28:1105–7. [14] Birkness KA, Swords WE, Huang PH, White EH, Dezzutti CS, Lal RB, et al. Observed differences in virulence-associated phenotypes between a human clinical isolate and a veterinary isolate of Mycobacterium avium. Infect Immun 1999;67:4895–901. [15] Han XY, Tarrand JJ, Infante R, Jacobson KL, Truong M. Clinical significance and epidemiologic analyses of Mycobacterium avium and Mycobacterium intracellulare among patients without AIDS. J Clin Microbiol 2005;43:4407–12. [16] Maekura R, Okuda Y, Hirotani A, Kitada S, Hiraga T, Yoshimura K, et al. Clinical and prognostic importance of serotyping Mycobacterium avium–Mycobacterium intracellulare complex isolates in human immunodeficiency virus-negative patients. J Clin Microbiol 2005;43:3150–8. [17] Smith I. Mycobacterium tuberculosis pathogenesis and molecular determinants of virulence. Clin Microbiol Rev 2003;16:463–96. [18] Danelishvili L, McGarvey J, Li YJ, Bermudez LE. Mycobacterium tuberculosis infection causes different levels of apoptosis and necrosis in human macrophages and alveolar epithelial cells. Cell Microbiol 2003;5:649–60. [19] Huttunen K, Jussila J, Hirvonen MR, Iivanainen E, Katila ML. Comparison of mycobacteria-induced cytotoxicity and inflammatory responses in human and mouse cell lines. Inhal Toxicol 2001;13:977–91. [20] Chan J, Xing Y, Magliozzo RS, Bloom BR. Killing of virulent Mycobacterium tuberculosis by reactive nitrogen intermediates produced by activated murine macrophages. J Exp Med 1992;175:1111–22. [21] Liu PT, Stenger S, Li H, Wenzel L, Tan BH, Krutzik SR, et al. Toll-like receptor triggering of a vitamin D-mediated human antimicrobial response. Science 2006;311:1770–3. [22] Thoma-Uszynski S, Stenger S, Takeuchi O, Ochoa MT, Engele M, Sieling PA, et al. Induction of direct antimicrobial activity through mammalian toll-like receptors. Science 2001;291:1544–7. [23] Abebe F, Mustafa T, Nerland AH, Bjune GA. Cytokine profile during latent and slowly progressive primary tuberculosis: a possible role for interleukin-15 in mediating clinical disease. Clin Exp Immunol 2006;143:180–92. [24] Turenne CY, Wallace Jr R, Behr MA. Mycobacterium avium in the postgenomic era. Clin Microbiol Rev 2007;20:205–29. [25] Sarmento AM, Appelberg R. Relationship between virulence of Mycobacterium avium strains and induction of tumor necrosis factor a production in infected mice and in in vitro-cultured mouse macrophages. Infect Immun 1995;63: 3759–64. [26] Nishiuchi Y, Kitada S, Maekura R. Liquid chromatography/mass spectrometry analysis of small-scale glycopeptidolipid preparations to identify serovars of Mycobacterium avium–intracellulare complex. J Appl Microbiol 2004;97: 738–48.
Contents lists available at ScienceDirect
Microbial Pathogenesis journal homepage: www.elsevier.com/locate/micpath
Virulence of Mycobacterium avium complex strains isolated from immunocompetent patients Yoshitaka Tateishi a, b, *, Yukio Hirayama a, Yuriko Ozeki a, c, Yukiko Nishiuchi a, d, Mamiko Yoshimura a, Jing Kang a, Atsushi Shibata a, Kazuto Hirata e, Seigo Kitada b, Ryoji Maekura b, Hisashi Ogura f, Kazuo Kobayashi g, Sohkichi Matsumoto a, * a
Department of Bacteriology, Osaka City University Graduate School of Medicine, 1-4-3 Asahi-machi, Abeno-ku, Osaka 545-8585, Japan Department of Internal Medicine, National Hospital Organization Toneyama National Hospital, 5-1-1 Toneyama, Toyonaka, Osaka 560-8552, Japan Sonoda Women’s University, 7-29-1 Minamitsukaguchi-cho, Amagasaki, Hyogo 661-8520, Japan d Toneyama Institute for Tuberculosis Research, Osaka City University Medical School, Toyonaka 5-1-1, Toneyama, Toyonaka, Osaka 560-8552, Japan e Department of Respiratory Medicine, Osaka City University Graduate School of Medicine, 1-4-3 Asahi-machi, Abeno-ku, Osaka 545-8585, Japan f Department of Virology, Osaka City University Graduate School of Medicine, 1-4-3 Asahi-machi, Abeno-ku, Osaka 545-8585, Japan g Department of Immunology, National Institute of Infectious Diseases, 1-23-1, Toyama, Shinjuku-ku, Tokyo 162-8640, Japan b c
a r t i c l e i n f o
a b s t r a c t
Article history: Received 6 August 2008 Received in revised form 29 September 2008 Accepted 2 October 2008 Available online 1 November 2008
Mycobacterium avium complex (MAC) disease has been increasing worldwide not only in immunocompromised but also in immunocompetent humans. However, the relationship between mycobacterial strain virulence and disease progression in immunocompetent humans is unclear. In this study, we isolated 6 strains from patients with pulmonary MAC disease. To explore the virulence, we examined the growth in human THP-1 macrophages and pathogenicity in C57BL/6 mice. We found that one strain, designated 198, which was isolated from a patient showing the most progressive disease, persisted in THP-1 cells. In addition, strain 198 grew to a high bacterial load with strong inflammation in mouse lungs and spleens 16 weeks after infection. To our knowledge, strain 198 is the first isolated MAC strain that exhibits hypervirulence consistently for the human patient, human macrophages in vitro, and even for immunocompetent mice. Other strains showed limited survival and weak virulence both in macrophages and in mice, uncorrelated to disease progression in human patients. We demonstrated that there is a hypervirulent clinical MAC strain whose experimental virulence corresponds to the serious disease progression in the patients. The existence of such strain suggests the involvement of bacterial virulence in the pathogenesis of pulmonary MAC disease in immunocompetent status. Ó 2008 Elsevier Ltd. All rights reserved.
Keywords: Mycobacterium avium complex Virulence Clinical isolates Immunocompetent humans Pulmonary disease
1. Introduction Mycobacterium avium complex (MAC) is the most common cause of human infection due to nontuberculous mycobacteria. Initially MAC was regarded as only an opportunistic pathogen, primarily in acquired immunodeficiency syndrome (AIDS) patients [1]; however, it has now been shown to cause progressive pulmonary disease even in immunocompetent humans [2]. The American Thoracic Society indicates a wide range of clinical manifestation in patients with non-AIDS MAC disease; some patients keep a stable condition for years, whereas others progress their illness rapidly [3]. Furthermore, MAC infection can be more difficult to treat
* Corresponding authors. Department of Bacteriology, Osaka City University, Graduate School of Medicine, 1-4-3 Asahi-machi, Abeno-ku, Osaka 545-8585, Japan. Tel.: þ81 6 6645 3746; fax: þ81 6 6645 3747. E-mail address: [email protected] (Y. Tateishi). 0882-4010/$ – see front matter Ó 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.micpath.2008.10.007
than M. tuberculosis due to even fewer available anti-microbial agents [3]. The pathogenesis of MAC infection has been recently investigated with respect to the host immune response. Interferongamma (IFN-g) activates macrophages to produce proteolytic enzymes and other metabolites, which exhibit mycobactericidal effects. Tumor necrosis factor-alpha (TNF-a), of which production is also stimulated by IFN-g, augments the bactericidal capacity of macrophages and plays a key role in the induction of the acquired immune response against mycobacteria [4]. A defective IFN-g response has been shown recently to cause disseminated MAC disease in IFN-g knock out mice and in humans with genetic mutations of IFN-g receptor [5,6] or autoantibodies to IFN-g in some young non-AIDS patients [7,8]. In addition to that, the activity of interleukin-10 (IL-10), which is known to inhibit cytokine synthesis by IFN-g-producing type1 helper T cells (Th1 cells), has been shown to increase susceptibility to MAC infection in immunocompetent mice [9].
Y. Tateishi et al. / Microbial Pathogenesis 46 (2009) 6–12
Besides genetic factors of the host, bacterial virulence should play an important role for the development of MAC disease. While isolates of M. tuberculosis are genetically homogeneous at the nucleotide level [10], MAC has high genetic diversity, including the presence of multiple plasmids [11], and thus likely to have a large corresponding diversity in virulence. In the most complete study examining virulence, forty-one MAC isolates from the environment as well as infected humans and animals were compared for virulence in C57BL/6 mice by intravenous injection [12]. Monitoring of the virulence by CFU counts in lungs, livers, and spleens over 4 months revealed three virulence phenotypes; high (logarithmically increasing load), intermediate (chronic infection at a constant load), and low (initial load increase followed by a decrease until clearance). In addition, clinical studies have suggested severe disease outcome in patients infected with some specific strain type of MAC. For example, MAC serovars 1, 4, and 8 Mycobacterium avium are associated with disease severity in AIDS patients [13], and a serovar 4 M. avium isolate from an AIDS patient was more invasive and proliferative in blood mononuclear cell-derived human macrophages than a serovar 2 strain from chickens [14]. In non-AIDS MAC disease, Mycobacterium intracellulare is associated with greater disease progression [15], and moreover, our previous prospective study on 68 non-AIDS patients suggests that serovar 4 M. avium is linked to greater disease progression with a pulmonary MAC infection [16]. Taking these previous data into consideration, we hypothesize that relatively hypervirulent MAC strains exist and may be associated with serious disease progression in immunocompetent patients. In order to elucidate the involvement of mycobacterial virulence in the pathogenesis of human pulmonary MAC disease, in this study we examined the difference of mycobacterial virulence of clinical isolates from patients with different disease types using human macrophages and immunocompetent mice.
7
and numerous bacteria in the sputum. The patient infected with strain 198 exhibited the most serious disease outcome among study patients in that a right pneumonectomy was needed to prevent disease progression. Strains 27, 36, and 347 were derived from patients with little progression of disease without chemotherapy (silent type). They exhibited lower levels of ESR, segmental pulmonary lesions in chest X-ray findings, and fewer bacteria in the sputum. The isolates belonging to the progressive type consisted of M. intracellulare unclassified serovar similar to serovar 12 (strain 198) and M. avium apolar type (strains 33 and 288). The isolates belonging to the silent type consisted of M. intracellulare serovar 1 (strain 27) and M. avium apolar type (strains 36 and 347). For comparison, we employed 2 veterinary strains of M. avium ATCC 25291 (serovar 2) as a highly virulent strain in mice [12] and ATCC 35767 (serovar 4) as a low virulent strain. Four clinical isolates other than strains 33 and 347, and ATCC 25291 formed the transparent colony morphology. Strain 33 produced both transparent and rough colony morphologies. Strain 347 and ATCC 35767 displayed smooth opaque colony morphology.
2.2. Growth of clinical isolates in 7H9 broth All strains showed logarithmic growth from 3 days after culture in 7H9 broth (Table 2). At day 5, two isolates from progressive type (strains 198 and 288) and one isolate from silent type (strain 36) grew significantly slower than ATCC 25291 (P < 0.005), and all clinical strains grew significantly slower than ATCC 35767 (P < 0.0001). The growth of strain 198 at day 5 was significantly slower than that of strain 27 (P ¼ 0.001), and was not significantly different from that of other clinical isolates.
2.3. Virulence of clinical isolates in THP-1 monocyte-derived macrophages
2. Results 2.1. Characteristics of mycobacterial strains Six clinical isolates of MAC were isolated from sputum of nonAIDS patients with pulmonary MAC disease, and designated 27, 33, 36, 198, 288, and 347 (Table 1). Strains 33, 198 and 288 were derived from patients with progressive disease against combination chemotherapy recommended by the American Thoracic Society guideline (progressive type) [3]. The patients with progressive disease exhibited higher levels of erythrocyte sedimentation rate (ESR), diffuse and severe pulmonary lesions in chest X-ray findings,
We next studied intracellular survival of the isolates. THP-1 cells, a human monocytic cell line, were differentiated into macrophages by treatment with phorbol 12-myristate 13-acetate (PMA) and infected with MAC strains. Strain 198 grew in THP-1 cells significantly higher than any other strains during 7 days of infection (P < 0.0001) (Table 3). Strain 198 grew to approximately 20-fold during 2 days of infection (P ¼ 0.005), and even at day 7, it kept the same level of bacterial load as day 0. Strain 36 also grew to approximately 2-fold during 2 days of infection (P ¼ 0.008); however, it was rapidly eliminated at day 7, similar to the other strains except for strain 198. There was no significant difference in
Table 1 Characteristics of isolated strains and clinical findings. Isolates
Species and serovar
Age
Sex
Duration of illness (years)
Erythrocyte sedimentation rate (mm/h)
Chest X-ray findingsa
Sputumb Smear
Culture
Progressive type 33 M. avium apolar type 198 M. intracellulare unclassified serovarc 288 M. avium apolar type
58 62 56
M F F
17 3 12
62 108 78
Advanced Advanced Advanced
2þ 2þ 2þ
3þ 2þ 2þ
Silent type 27 36 347
67 54 79
F F F
17 9 14
50 29 50
Moderate Moderate Moderate
– – 1þ
1þ 1þ 1þ
M. intracellulare serovar 1 M. avium apolar type M. avium apolar type
Data and sputum samples were collected at the enrollment of the study in 2003. a Advanced chest X-ray findings were defined as bilateral cavities, giant cavities, or bilateral bronchiectasis, and moderate findings were defined as focal inflammation, small or fewer cavities, or mild bronchiectasis. b Smear findings of sputum were defined as follows in high performance fields of microscopy; -: no bacteria in all fields, 1þ: less than one bacteria in several fields, 2þ: approximately 1–12 bacteria in one field. Culture findings were defined as follows using Ogawa egg agar; 1þ: colonies less than 200, 2þ: colonies more than 200 and less than 500, 3þ: colonies more than 500 and less than 2000. c The serovar of strain 198 was identified as a new type similar to serovar 12 determined by the liquid chromatography/mass spectrometry.
8
Y. Tateishi et al. / Microbial Pathogenesis 46 (2009) 6–12
Table 2 Growth rate of MAC in 7H9 broth. Strain
33 198 288 27 36 347 25291 35767
Ratio of CFUs ata Day 1
Day 3
Day 5
0.91 0.28 0.93 0.17 0.85 0.20 1.1 0.25 0.83 0.093 0.96 0.16 1.6 0.25 0.97 0.12
38 0.86 12 1.9 4.3 1.9 7.3 2.2 5.4 0.21 2.7 1.1 4.8 0.24 11 3.0
63 10 18 4.0 16 0.95 120 21 11.0 1.7 44 19 110 16* 370 43**
and spleens, and severe pulmonary inflammation at 16 weeks of infection. By contrast, other clinical isolates did not increase profoundly in lung CFUs; however, these strains were never eliminated from lungs. ATCC 35767 was rapidly decreased and undetectable in lungs, spleens and livers within 16 weeks of infection. Overall, the clinical isolates other than strain 198 exhibited limited histological lesions with transient inflammatory changes in lungs 4 weeks after the inoculation, and thereafter the inflammation subsided at 16 weeks. 3. Discussion
*
Significantly different (P < 0.005) from values for strain 198, 288, and 36 as calculated by Scheffe´’s test. ** Significantly different (P < 0.0001) from values for all clinical strains as calculated by Scheffe´’s test. a Means standard deviations of the ratio of CFUs to those at day 0.
the growth rate among these strains except strain 198 during infection. On light microscopic observation, THP-1 cell morphologies were not different between infected and uninfected cells (data not shown). We then assessed cytotoxicity by the levels of lactate dehydrogenase (LDH) released into the culture supernatants at day 7. The LDH release was detectable in strain 33 and the laboratory strains (strain 33; 5.8 1.5%, ATCC 25291; 11 1.0%, ATCC 35767; 12 1.9%, without significant difference among these strains); however, it was not detectable in other clinical isolates.
2.4. Pathogenesis of clinical isolates in mice Female C57BL/6 mice were infected by intratracheal instillation with each strain. Bacterial load in lungs, livers, and spleens were evaluated, and histological inflammation was visually analyzed in 5-mice per strain at defined time points during 16 weeks of infection. There was no significant difference in lung CFUs among strains tested 1 day after the inoculation. Strain 198 showed high bacterial load, and tended to increase gradually both in lungs and spleens during 16 weeks of infection (P ¼ 0.08 between day 1 and 16 weeks) (Fig. 1). Strain 198 was loaded in lungs significantly higher than strain 27 (P ¼ 0.04) and 33 (P ¼ 0.0006) at 8 weeks of infection, and than strain 33 (P ¼ 0.0009), 288 (P ¼ 0.001), 36 (P ¼ 0.0003), and 347 (P ¼ 0.004) at 16 weeks. Histologically, strain 198 induced strong inflammation in lungs, which was paralleled with bacterial loads (Fig. 2). ATCC 25291, known as highly a virulent strain in mice [12], showed initial reduction of bacterial load in lungs at 4 weeks of infection (P ¼ 0.01 between day 1 and 4 weeks) and rapid increase in bacterial load in lungs after 4 weeks of infection. ATCC 25291 was comparatively virulent to strain 198 with respect to the high bacterial load in lungs
Table 3 Growth rate of MAC in THP-1 cells. strain
33 198 288 27 36 347 25291 35767 *
Ratio of CFUs ata Day 2
Day 7
0.29 0.12 18 12* 0.47 0.26 0.54 0.37 2.4 1.2 1.1 0.49 0.16 0.048 0.059 0.029
0.25 0.16 1.30 0.68* 0.097 0.055 0.28 0.11 0.31 0.11 0.36 0.23 0.11 0.038 0.0070 0.0048
Significantly different (P < 0.0001) from values for any other strains studied as calculated by Scheffe´’s test. a Means standard deviations of the ratio of CFUs to those at day 0.
Virulence is defined as the quantitative ability of an agent to cause disease. The virulence of mycobacteria can be evaluated by the infection to macrophages and animals [17]. This is the first study that examined the virulence of MAC isolates from immunocompetent patients with different types of disease outcome. We found that strain 198, which derived from a patient with most serious disease, revealed high bacterial load both in THP-1 cells and in C57BL/6 mice among isolates studied. Strain-specific virulence of MAC has been implicated by some previous studies of the serovar 4 M. avium isolated from patients. In AIDS-related MAC disease, a serovar 4 M. avium isolate has shown to be one of the frequently isolated type [13], and a previous analysis of a serovar 4 isolate and ATCC strain has shown the superior virulence of serovar 4 M. avium in human macrophages [14]. In non-AIDS pulmonary MAC disease, our recent prospective study indicates that patients infected with serovar 4 M. avium has poorer prognosis than those infected with MAC of other serovars [16]; however, to our best knowledge, no study has shown the direct data of mycobacterial virulence of clinical isolates and clinical disease outcome. Strain 198 is the first MAC isolate whose experimental virulence corresponds to the serious disease outcome in humans. Thus, strain 198 has strainspecific strong virulence for immunocompetent humans and mice. We consider that strain 198 is worth further genetic investigation of virulence factors. MAC strains hypervirulent for mice has been isolated previously by Pedrosa et al. including ATCC 25291 and MAC 101, which proliferate profoundly in mouse macrophages and in mice in vivo [12]. In this study, strain 198 proliferated in human macrophages, in correspondence with rapid clinical disease progression and additionally in mouse lungs. The consistency between experimental virulence in human cells and clinical disease outcome suggests that the capability of inducing such strong pathogenesis may be attributed mostly to the characteristics of the pathogen, i.e. virulence factor(s) for mammalian cells unique to strain 198. Previously Birkness et al. has shown the strong cytotoxic effect and growth of serovar 4 M. avium isolated from an AIDS patient in blood mononuclear cell-derived human macrophages compared with a serovar 2 strain from chickens (ATCC 35713). Therefore, we evaluated cytotoxicity of MAC strains by the microscopic morphology and by the LDH release from infected THP-1 cells; however, contrary to the expectation, strain 198 was not cytotoxic to THP-1. In addition, the release of LDH was lower in strain 33, ATCC 25291, and ATCC 35767 than in the previous experiment of M. tuberculosis infection to THP1 cells (cytotoxicity in cases of M. tuberculosis H37Rv and H37Ra; approximately 30%) [18]. We assume that cytotoxic effect may not play a major role in displaying the virulence of MAC during infection, suggested by the similar result by Huttunen et al. showing the lack of cytotoxic effect of MAC in human 28SC macrophage and A549 lung epithelial cell lines evaluated by 3-(4,5-dimethyl-2thiazolyl)-2,5-diphenyl-2H tetrazolium bromide (MTT) assay [19]. We speculate the virulence of strain 198 depends on the ability to survive or proliferate in macrophages rather than cytotoxic effect, which, may be causative for severe pulmonary MAC disease with rapid disease progression within a few years.
Y. Tateishi et al. / Microbial Pathogenesis 46 (2009) 6–12
CFUs/organ
109
33
109 108
108
107
107
107
106
106
106
105
105
105
104
104
104
103
103
103
102
102
102
101
101
101
109
CFUs/organ
109
108
1day 4wks 8wks 16wks
1day 4wks 8wks 16wks
27
109
36
109
108
108
107
107
107
106
106
106
105
105
105
104
104
104
103
103
103
102
102
102
101
101
101
109
25291
1day 4wks 8wks 16wks 109
108
108
107
107
106
106
105
105
104
104
103
103
102
102
101
101 1day 4wks 8wks 16wks
288
1day 4wks 8wks 16wks
108
1day 4wks 8wks 16wks
CFUs/organ
198
9
347
1day 4wks 8wks 16wks
35767 Lung Liver Spleen
1day 4wks 8wks 16wks
Fig. 1. Time course of mycobacterial growth in lungs, spleens and livers of C57BL/6 mice. Bacterial suspensions containing 1 105 CFUs were inoculated intratracheally to female C57BL/6 mice at the age of 7 weeks (n ¼ 20 per strain). The lungs, livers and spleens of 5 mice per strain were sectioned at day 1 (only lungs), 4, 8, and 16 weeks later from challenge Data were presented as means standard deviations of CFUs/organ.
In this study, strain 198 showed strong virulence for mice at 16 weeks of infection similar to ATCC 25291; however, virulence for THP-1 cells was quite different, and pathogenic effects for mice within 4 weeks of infection was dissimilar between these two strains. These differences can be explained by the difference of immune response between host species and by the difference of immune phase. First, strain 198 could proliferate, but ATCC 25291 was rapidly eliminated in THP-1 cells (Table 3). In mouse macrophages mycobactericidal activity is attributed to nitric oxide produced by inducible nitric oxide synthase [20], whereas in human macrophages, it is attributed to Toll-like receptor signalingdependent production of anti-microbial peptides [21,22]. Strain 198 is capable of proliferating under these two patterns of mycobactericidal activities, which suggests that strain 198 may have some virulence factors advantageous to survive both in human and
mouse macrophages against mycobactericidal activity of the hosts, in contrast to ATCC 25291 which may lack virulence factors to survive in human macrophages. Second, strain 198 showed high bacterial load in lungs continuingly during 16 weeks of infection in mice, while ATCC 25291 proliferated after initial reduction in lungs at 4 weeks of infection (Fig. 1). The in vitro infection model using cell lines and the in vivo infection model using mice within 4 weeks reflects early stages of infection; on the other hand, the in vivo model after 8 weeks reflects chronic phase of infection [17,23]. The difference of pathogenic effects for mice within 4 weeks of infection suggests that strain 198 may resist both innate and acquired immunity, while ATCC 25291 may resist acquired immunity only. We assume that strain 198 and ATCC 25291 may possess different virulence mechanisms to persist in in vivo after development of acquired immunity.
10
Y. Tateishi et al. / Microbial Pathogenesis 46 (2009) 6–12
Fig. 2. Histological pictures of the lungs during 4 weeks or 16 weeks of infection in C57BL/6 mice by hematoxylin–eosin staining. Magnification, 40.
In this study, clinical strains except for strain 198 did not show consistent virulence-associated phenotype among THP-1 cells, C57BL/6 mice, and clinical disease outcome. Similarly, Pedrosa et al. has also revealed that the growth of MAC in bone-marrow derived macrophages does not necessarily predict the virulence in mice by comparing the growth of 41 MAC isolates from various derives including humans, animals, and environment [12]. Although some clinical cases of pulmonary MAC disease may be caused by hypervirulent strains such as strain 198, these findings of clinical and
natural isolates suggest that virulence may not the only determinant of the pathogenesis of pulmonary MAC disease in the majority of clinical cases. The development of pulmonary MAC disease depends on the balance between bacterial virulence and host defense. It is widely accepted that patients with pulmonary MAC disease have some characteristics of clinical background, such as males in their 40s and early 50s who have a history of cigarette smoking and excessive alcohol use, and such as postmenopausal, nonsmoking females [3], and these patient characteristics might
Y. Tateishi et al. / Microbial Pathogenesis 46 (2009) 6–12
possibly indicate unknown predisposing conditions which enhance susceptibility for pulmonary MAC infection. The diverse phenotype of clinical MAC strains may be attributed to the disease susceptibility of the hosts. We propose that pathogenic mechanism of human pulmonary MAC disease include two patterns; one is that the strong virulence of MAC strains such as strain 198 induces rapid mycobacterial growth and serious disease outcome, and the other is that relatively weak to moderate virulence interacts with predisposing conditions of the host, leading the wide range of clinical outcome. This study was preliminary in that we did not identify the mechanism of hypervirulence of strain 198. We observed the consistency between hypervirulence in human macrophages bedsides in immunocompetent mice and severe clinical outcome only in strain 198, not in any other isolates studied. From this finding, we speculate the existence of strain-specific virulence factors of strain 198. Recent exponential advances have enabled whole genome sequence of two M. avium strains, M. avium 104 and M. avium subsp. paratuberculosis K-10. Based on these exhaustive information, comparative genomics of MAC organisms has revealed the different genomic components regarding virulence factors, such as ser2 encoding glycosylation enzyme of the lipopeptide core to generate the glycopeptidolipids, mammalian cell entry (mce) gene homologs, and PE/PPE genes (i.e., with Pro Glu and Pro Pro Glu motifs) [24]. In addition, there are large sequence polymorphisms among MAC organisms, suggesting a large corresponding diversity in virulence [11,24]. We speculate that the virulence of MAC strains including strain 198 may be determined by insertion or deletion of virulence genes encoding known [24] or unknown virulence factors. In summary, we demonstrated that certain clinical strain derived from patients of the progressive pulmonary MAC disease exhibits strong virulence in human macrophages and in immunocompetent mice. Among clinical isolates, strain 198 is the first isolate hypervirulent to both human macrophages and mice. Our data suggest that strain-to-strain differences in virulence may play a significant role in disease progression in humans. Although Sarmento et al. showed that capability of TNF-a production from macrophages inversely correlates with the virulence of MAC strains [25], we could not find such relationship among the isolates (data not shown). In future studies, we will identify the virulence/pathogenicity-associated factor(s) of strain 198 and survey the frequency of strain variation in immunocompetent patients with pulmonary MAC disease. 4. Materials and methods 4.1. Bacterial strains We used six clinical isolates from non-AIDS patients with pulmonary MAC disease and two laboratory strains, M. avium ATCC 25291 (serovar 2) and M. avium ATCC 35767 (serovar 4), in this study. Clinical isolates were obtained between September and November in 2003 at Toneyama National Hospital. Informed consent was obtained from all patients according to the guideline of Institutional Review Board of Toneyama National Hospital. Diagnosis of pulmonary MAC disease was made according to the American Thoracic Society guideline [3]. The samples were derived from two groups of patients; one group exhibited progressive disease in spite of the combination chemotherapy including clarithromycin, ethambutol and rifampin recommended by the American Thoracic Society guideline (progressive type) [3], the other displayed no exacerbation without anti-microbial chemotherapy for approximately ten years or more (silent type). These types were determined by the laboratory findings at the period of sputum sampling (including sputum smear and culture,
11
chest X-ray findings, and erythrocyte sedimentation rate) and the rapidness of disease progression (Table 1). Sputum specimens were mixed with 2% sodium hydroxide, and N-acetyl-L-cysteine and then centrifuged for 15 min at 3000 g. The supernatants were discarded, and the sediment was mixed at 1:10 (vol/vol) with sterile water. The bacteria were cultivated in Middlebrook 7H9 broth supplemented with albumin–dextrose–catalase, 0.02% glycerin and 0.05% Tween 80, and then kept at 80 C until following experiments. Identification of MAC was made by polymerase chain reaction using a commercially available kit (AMPLICOR Mycobacterium Tuberculosis Test, Roche, Basel, Switzerland). The serovars of clinical isolates were identified by the liquid chromatography/mass spectrometry as described previously [26]. Strains not containing serovar-specific oligosaccharides were defined as apolar type.
4.2. Growth in 7H9 broth Bacterial suspension was adjusted to be 0.2 by optical density (OD) at 630 nm. The samples were cultured in 5 ml of 7H9 media in plastic tubes without agitation After vortexing to dissolve aggregates, cultivated bacterial suspensions were inoculated at days 1, 3, and 5 by serial 10-fold dilutions on Middlebrook 7H11 agar plates supplemented with oleic acid–albumin–dextrose–catalase, and 0.05% glycerol (7H11-OADC) agar plates in triplicate. The number of CFUs was counted after cultivating at 37 C for 3 weeks. 4.3. Infection of THP-1 cells with MAC in vitro THP-1 cells were purchased from Health Science Research Resources Bank (Tokyo, Japan). The cells were cultured in RPMI1640 containing 10% heat-inactivated fetal bovine serum (FBS; Equitech-bio, TX), and subcultured every 3–4 days. THP-1 cells were differentiated by 100 nM PMA (Sigma–Aldrich, St Louis, MO) for 48 h before infection. Before 48 h of infection, 1 ml of 2 105/ml cells was cultured in RPMI1640 containing 5% human serum (ABblood group) in 24-well plates. Then, 1 ml of 2 104 CFUs/ml bacteria was exposed to the cultured cells for 24 h without opsonization (multiplicity of infection; 0.1 bacteria/cell). After that, the cells were treated with 20 mg/ml of gentamicin for 3 h to kill extracellular bacteria, followed by washing 4 times by RPMI1640. The infected cells were cultured in 2 ml of RPMI1640 containing 5% human serum. At days 0 and 7, uninfected bacteria were removed by washing with RPMI1640 4 times, and 500 ml of filter-sterilized phosphate buffered saline containing 0.5% Triton X-100 (Wako, Osaka, Japan) was treated per well to lyse cell membrane. The intracellular survival of bacteria was determined by counting CFUs by inoculating the cell lysate on 7H11-OADC agar plates. The experiment was performed in triplicate.
4.4. Assays for cytotoxicity Cytotoxic effects were evaluated by the release of LDH from the cells. LDH activity of culture supernatants was determined by a commercially available kit (Roche, Basel, Switzerland). Supernatants were diluted to be 101 by distilled water for optimal reaction. The diluents were reacted with reaction mixture for 30 min, and then the OD was measured at 492 nm. Supernatants of completely lysed uninfected cells with filter-sterilized phosphate buffered saline containing 20% Triton X-100 and those of uninfected cells untreated with Triton X-100 were served as high and low controls, respectively. Cytotoxicity (%) was calculated as follows; ðODsample ODlow control Þ 100=ðODhigh control ODlow control Þ. The measurement was performed in triplicate.
12
Y. Tateishi et al. / Microbial Pathogenesis 46 (2009) 6–12
4.5. Animal studies Female C57BL/6 mice aged at 6 weeks were purchased from CLEA Japan (Tokyo, Japan). All mice were kept under specific pathogen free conditions in animal facility of Osaka City University Graduate School of Medicine according to the institutional guidelines for the animal experiments. Twenty mice were used per group for infecting with each strain. One hundred microlitre of bacterial suspension containing 1 105 CFUs of MAC was inoculated into the trachea of the 7 weeks-aged mice anesthetized with pentobarbital sodium. Lungs, spleens and livers were removed on day 1 (only lungs) and 4, 8, 16 weeks after inoculation from 5-mice per strain. The organs were homogenized in 1 ml saline, and 0.1 ml of 10-fold dilutions of the homogenates was plated on 7H11-OADC agar followed by cultivating for 3 weeks. Bacterial burden was evaluated by CFUs per organ. Histological sections were made by standard methods including formalin fixation, dehydration, embedding in paraffin, and staining with hematoxylin and eosin. 4.6. Statistical analysis Data were analyzed using the statistical analysis software package StatView 5.0 (SAS Institute, Cary, NC). The difference of mycobacterial growth in 7H9 broth, THP-1 cells, and mice was compared by a post hoc test of Scheffe´ among the strains tested. The difference of mycobacterial growth at defined time points during infection in THP-1 cells as well as in mice was compared by repeated measurement ANOVA with a post hoc test of Scheffe´ in the individual strains. Difference was considered statistically significant at P < 0.05. Acknowledgement We thank Todd P. Primm for the critical comments on the manuscript. We also thank Chihiro Inoue and Sara Matsumoto for the assistance of the experiments and for the heartfelt encouragement. This work was supported by grants from the Ministry of Education, Culture, Sports, Science and Technology, the Ministry of Health, Labour and Welfare (Research on Emerging and Reemerging Infectious Diseases, Health Sciences Research Grants), The Japan Health Sciences Foundation, and The United States-Japan Cooperative Medical Science Program against Tuberculosis and Leprosy. References [1] Horsburgh Jr CR, Gettings J, Alexander LN, Lennox JL. Disseminated Mycobacterium avium complex disease among patients infected with human immunodeficiency virus, 1985–2000. Clin Infect Dis 2001;33:1938–43. [2] Field SK, Fisher D, Cowie RL. Mycobacterium avium complex pulmonary disease in patients without HIV infection. Chest 2004;126:566–81. [3] Griffith DE, Aksamit T, Brown-Elliott BA, Catanzaro A, Daley C, Gordin F, et al. An official ATS/IDSA statement: diagnosis, treatment, and prevention of nontuberculous mycobacterial diseases. Am J Respir Crit Care Med 2007;175: 367–416.
[4] Flynn JL, Goldstein MM, Chan J, Triebold KJ, Pfeffer K, Lowenstein CJ, et al. Tumor necrosis factor-a is required in the protective immune response against Mycobacterium tuberculosis in mice. Immunity 1995;2:561–72. [5] Dorman SE, Picard C, Lammas D, Heyne K, van Dissel JT, Baretto R, et al. Clinical features of dominant and recessive interferon g receptor 1 deficiencies. Lancet 2004;364:2113–21. [6] Newport MJ, Huxley CM, Huston S, Hawrylowicz CM, Oostra BA, Williamson R, et al. A mutation in the interferon-g-receptor gene and susceptibility to mycobacterial infection. N Engl J Med 1996;335:1941–9. [7] Kampmann B, Hemingway C, Stephens A, Davidson R, Goodsall A, Anderson S, et al. Acquired predisposition to mycobacterial disease due to autoantibodies to IFN-g. J Clin Invest 2005;115:2480–8. [8] Patel SY, Ding L, Brown MR, Lantz L, Gay T, Cohen S, et al. Anti-IFN-g autoantibodies in disseminated nontuberculous mycobacterial infections. J Immunol 2005;175:4769–76. [9] Roque S, Nobrega C, Appelberg R, Correia-Neves M. IL-10 underlies distinct susceptibility of BALB/c and C57BL/6 mice to Mycobacterium avium infection and influences efficacy of antibiotic therapy. J Immunol 2007;178:8028–35. [10] Ernst JD, Trevejo-Nunez G, Banaiee N. Genomics and the evolution, pathogenesis, and diagnosis of tuberculosis. J Clin Invest 2007;117:1738–45. [11] Primm TP, Lucero CA, Falkinham 3rd JO. Health impacts of environmental mycobacteria. Clin Microbiol Rev 2004;17:98–106. [12] Pedrosa J, Florido M, Kunze ZM, Castro AG, Portaels F, McFadden J, et al. Characterization of the virulence of Mycobacterium avium complex (MAC) isolates in mice. Clin Exp Immunol 1994;98:210–6. [13] Hoffner SE, Kallenius G, Petrini B, Brennan PJ, Tsang AY. Serovars of Mycobacterium avium complex isolated from patients in Sweden. J Clin Microbiol 1990;28:1105–7. [14] Birkness KA, Swords WE, Huang PH, White EH, Dezzutti CS, Lal RB, et al. Observed differences in virulence-associated phenotypes between a human clinical isolate and a veterinary isolate of Mycobacterium avium. Infect Immun 1999;67:4895–901. [15] Han XY, Tarrand JJ, Infante R, Jacobson KL, Truong M. Clinical significance and epidemiologic analyses of Mycobacterium avium and Mycobacterium intracellulare among patients without AIDS. J Clin Microbiol 2005;43:4407–12. [16] Maekura R, Okuda Y, Hirotani A, Kitada S, Hiraga T, Yoshimura K, et al. Clinical and prognostic importance of serotyping Mycobacterium avium–Mycobacterium intracellulare complex isolates in human immunodeficiency virus-negative patients. J Clin Microbiol 2005;43:3150–8. [17] Smith I. Mycobacterium tuberculosis pathogenesis and molecular determinants of virulence. Clin Microbiol Rev 2003;16:463–96. [18] Danelishvili L, McGarvey J, Li YJ, Bermudez LE. Mycobacterium tuberculosis infection causes different levels of apoptosis and necrosis in human macrophages and alveolar epithelial cells. Cell Microbiol 2003;5:649–60. [19] Huttunen K, Jussila J, Hirvonen MR, Iivanainen E, Katila ML. Comparison of mycobacteria-induced cytotoxicity and inflammatory responses in human and mouse cell lines. Inhal Toxicol 2001;13:977–91. [20] Chan J, Xing Y, Magliozzo RS, Bloom BR. Killing of virulent Mycobacterium tuberculosis by reactive nitrogen intermediates produced by activated murine macrophages. J Exp Med 1992;175:1111–22. [21] Liu PT, Stenger S, Li H, Wenzel L, Tan BH, Krutzik SR, et al. Toll-like receptor triggering of a vitamin D-mediated human antimicrobial response. Science 2006;311:1770–3. [22] Thoma-Uszynski S, Stenger S, Takeuchi O, Ochoa MT, Engele M, Sieling PA, et al. Induction of direct antimicrobial activity through mammalian toll-like receptors. Science 2001;291:1544–7. [23] Abebe F, Mustafa T, Nerland AH, Bjune GA. Cytokine profile during latent and slowly progressive primary tuberculosis: a possible role for interleukin-15 in mediating clinical disease. Clin Exp Immunol 2006;143:180–92. [24] Turenne CY, Wallace Jr R, Behr MA. Mycobacterium avium in the postgenomic era. Clin Microbiol Rev 2007;20:205–29. [25] Sarmento AM, Appelberg R. Relationship between virulence of Mycobacterium avium strains and induction of tumor necrosis factor a production in infected mice and in in vitro-cultured mouse macrophages. Infect Immun 1995;63: 3759–64. [26] Nishiuchi Y, Kitada S, Maekura R. Liquid chromatography/mass spectrometry analysis of small-scale glycopeptidolipid preparations to identify serovars of Mycobacterium avium–intracellulare complex. J Appl Microbiol 2004;97: 738–48.